Comparative Biochemistry and Physiology, Part A 203 (2017) 69–76

Contents lists available at ScienceDirect

Comparative Biochemistry and Physiology, Part A

journal homepage: www.elsevier.com/locate/cbpa

Rumen content stratification in the giraffe (Giraffa camelopardalis)

Cathrine Sauer a,b,1, Marcus Clauss c,⁎, Mads F. Bertelsen b, Martin R. Weisbjerg a, Peter Lund a
a
Department of Animal Science, AU Foulum, Aarhus University, Blichers Allé 20, PO Box 50, DK-8830 Tjele, Denmark
b
Center for Zoo and Wild Animal Health, Copenhagen Zoo, Roskildevej 32, DK-2000 Frederiksberg, Denmark
c
Clinic for Zoo Animals, Exotic Pets and Wildlife, Vetsuisse Faculty, University of Zurich, Winterthurerstr. 260, CH-8057 Zurich, Switzerland

a r t i c l e i n f o a b s t r a c t

Article history: Ruminants differ in the degree of rumen content stratification, with ‘cattle-types’ (i.e., the grazing and
Received 5 April 2016 intermediate feeding ruminants) having stratified content, whereas ‘moose-types’ (i.e., the browsing ruminants)
Received in revised form 30 August 2016 have unstratified content. The feeding ecology, as well as the digestive morphophysiology of the giraffe (Giraffa
Accepted 30 August 2016
camelopardalis), suggest that it is a ‘moose-type’ ruminant. Correspondingly, the giraffe should have an
Available online 01 September 2016
unstratified rumen content and an even rumen papillation pattern. Digesta samples were collected from along
Keywords:
the digestive tract of 27 wild-caught giraffes kept in bomas for up to 2 months, and 10 giraffes kept in zoological
Ruminant gardens throughout their lives. Samples were analysed for concentration of dry matter, fibre fractions, volatile
Browser fatty acids and NH3, as well as mean particle size and pH. There was no difference between the dorsal and ventral
Digestion rumen region in any of these parameters, indicating homogenous rumen content in the giraffes. In addition to the
Fermentation digesta samples, samples of dorsal rumen, ventral rumen and atrium ruminis mucosa were collected and the
Feeding type papillary surface enlargement factor was determined, as a proxy for content stratification. The even rumen
Anatomy papillation pattern observed also supported the concept of an unstratified rumen content in giraffes. Zoo giraffes
Physiology
had a slightly more uneven papillation pattern than boma giraffes. This finding could not be matched by
differences in physical characteristics of the rumen content, probably due to an influence of fasting time ante
mortem on these parameters.
© 2016 Elsevier Inc. All rights reserved.

1. Introduction reticulorumen and classifies ruminant digestive tracts as either

The stratification of rumen content is a hallmark of domestic rumi-
nant digestive physiology (Hummel et al., 2009). In the original research
on anatomical differences between browsing and grazing ruminants
(Hofmann, 1973), the absence of a distinct rumen content stratification
in browsing ruminants was a side finding (Hofmann, 1989; Renecker
and Hudson, 1990). The dichotomy between stratified rumen content
in grazers and unstratified content in browsers was subsequently
emphasized as a major physiological difference (Clauss et al., 2003),
and anatomical differences between the feeding types were interpreted
based on this characteristic. The concept that adaptations served to
reinforce the presence or absence of stratification (Clauss et al., 2008)
was modified after feeding experiments showed that the one
mechanism most intuitively linked to stratification - the particle sorting
mechanism - works in a very similar way in ruminants with and with-
out rumen content stratification (Lechner et al., 2010). A newer concept
considers the presence or absence of rumen content stratification as a
consequence of either a low or high fluid throughput through the
⁎ Corresponding author.
E-mail address: mclauss@vetclinics.uzh.ch (M. Clauss).
1
Present address: Chester Zoo, Caughall Road, Upton-by-Chester, Chester CH2 1LH,
United Kingdom.
‘moose-type’ or ‘cattle-type’ (Clauss et al., 2010). ‘Moose-type’ rumi-
nants have unstratified rumen contents and a low fluid throughput,
possibly because they are rate-limited in the production of viscous sali-
va with tannin-binding proteins (Hofmann et al., 2008), and are strict
browsers. ‘Cattle-type’ ruminants have a fluid throughput through the
forestomach that is faster than that of ingesta particles, stratified
rumen contents, and typically include grass in their natural diet,
i.e., are intermediate feeders or grazers (Codron and Clauss, 2010).
The current hypothesis is that the evolutionary driver for a ‘cattle-
type’ physiology is the increased microbial yield and increased microbi-
al efficiency facilitated by an increased harvest of microbes by the
continuous ‘washing’ of forestomach contents (Dittmann et al., 2015;
Hummel et al., 2015), making the stratification of rumen contents a
mere side effect.
Rumen content stratification can be determined by various methods.
In live animals, ultrasound examination can distinguish between strati-
fied content with a dorsal gas dome and unstratified content without
such a dome (Tschuor and Clauss, 2008). In fistulated animals, sampling
of digesta from different regions of the rumen can demonstrate differ-
ences in dry matter (DM) content, particle size, and other measure-
ments of fermentation processes indicative of content stratification
(Hummel et al., 2009; Lechner et al., 2010). Other quantitative methods
have been applied as well, such as a device that measured the resistance

http://dx.doi.org/10.1016/j.cbpa.2016.08.033
1095-6433/© 2016 Elsevier Inc. All rights reserved.
70 C. Sauer et al. / Comparative Biochemistry and Physiology, Part A 203 (2017) 69–76
met by a weight pulled through the different stratification layers to
determine “ingesta consistency” in fistulated animals (Welch, 1982).
Differences in rumen content characteristics indicating stratification
are difficult to demonstrate in dead animals, as there is a risk of mixing
the rumen content during carcass handling. The stratification is main-
tained if carcasses are frozen in the natural resting position (Clauss
et al., 2016). Careful handling and dissection of the fresh carcass,
i.e., maintaining the natural position as much as possible, does allow
sampling of the forestomachs in a way that differences between sam-
pling locations can be assessed in terms of DM content and particle
size (Clauss et al., 2009a) or pH (Ritz et al., 2013). Even if careful sam-
pling of forestomach content to identify the degree of stratification is
not an option, content stratification is still reflected in the rumen
papillation pattern (Clauss et al., 2009b), as papillae growth is stimulat-
ed by the presence of volatile fatty acids and thus represents an
integrated measure of content stratification over the last months. In
contrast, rumen content characteristics will by nature only reflect the
last few meals.
Characteristics of the digestive tract anatomy suggest that the giraffe
(Giraffa camelopardalis) is a ‘moose-type’ ruminant (Sauer et al., 2016).
This matches observations in free-ranging animals that are strict
browsers (Leuthold and Leuthold, 1972; Pellew, 1984) that only very
rarely consume grass (Seeber et al., 2012).
Browsing animals, such as the giraffe, are typically considered more
difficult to maintain in captivity as compared to grazing animals (Müller
et al., 2011). In captivity, browsing ruminants generally excrete larger
faecal particles than grazing ruminants (Clauss et al., 2002). In contrast,
no such difference is evident when comparing free-ranging grazers and
browsers (Hummel et al., 2008; Lechner et al., 2010). One possible inter-
pretation of these findings is that whereas the teeth of both feeding
types are adapted to their respective natural diets (Heywood, 2010;
Kaiser et al., 2010), those of browsers are less suited to comminute
the food they receive in captivity. This is potentially exacerbated by
the excessive tooth wear experienced by giraffes in captivity (Clauss
et al., 2007). In giraffes, inadequate chewing efficiency might explain
the apparently high prevalence of digestive tract phytobezoars
(conglomerates of plant particles) in captive animals (Hummel and
Clauss, 2006). Another indicator of sub-optimal diets in captivity was
the drastic reduction in both number and size of rumen papillae
observed in two captive giraffes by Hofmann and Matern (1988).
In this study, we investigated the rumen contents stratification in
giraffes kept in zoos or in short-term boma confinement. Boma giraffes
were expected to have relatively homogenous rumen content with little
difference between the dorsal and ventral rumen in terms of DM
concentration, particle size, pH and measurements of fermentation.
This absence of stratification was assumed to be matched by a homoge-
nous rumen papillation pattern. Given reports of differences between
free-ranging and captive animals (Hofmann and Matern, 1988), we
hypothesized that giraffes kept and raised in zoos would have more
stratified rumen contents and a less even rumen papillation pattern,
as well as larger digesta particle sizes, than giraffes recently caught
from the wild.
2. Materials and methods
Samples were collected from 27 giraffes (25 males and 2 females)
caught from the wild and kept in bomas, ranging in body mass
from 280 to 660 kg (mean ± SD: 468 ± 95 kg), and 10 zoo giraffes
(6 males and 4 females) ranging in body mass from 182 to 1225 kg
(mean: 703 ± 272 kg), though not all samples were collected from
every individual due to practical and time limitations.
Wild giraffes were caught in South Africa or Namibia and housed in
bomas by Wildlife Assignments International Ltd., Hammanskraal,
South Africa, for approximately 2 months prior to euthanasia and dis-
section. During this period, the giraffes were group housed and
received a diet of fresh locally cut savannah browse, leafy lucerne hay
(Medicago sativa), and Boskos pellets (based on ground native trees
and various protein, energy and mineral sources, Wes Enterprises Ltd.,
South Africa). Water was freely available. All giraffe consumed food dur-
ing the stay in the boma, but it was not possible to quantify individual
intake. With permission from the Gauteng Province of South Africa,
the boma giraffes were euthanized following various physiological ex-
periments conducted by the Danish Cardiovascular Giraffe Research
project (e.g. Smerup et al., 2016). Due to the experimental procedures,
many boma giraffes were fasted overnight prior to anaesthesia and sub-
sequent euthanasia. The duration of the fasting time was estimated
based on when the giraffes last had access to feed and the time of eutha-
nasia. The estimated fasting time for boma giraffes ranged from 0 to 48 h
with a mean duration of 18 ± 11 h.
Captive giraffes from six Danish and one Swedish zoo were culled for
management reasons or because of chronic osteoarthritis. These
animals were all group fed on diets consisting of hay (in 6 cases lucerne
hay only, in 3 cases grass hay only, and one animal had access to both
hay types), various concentrate pellets and as much browse as possible.
Limited amounts of other feeds including Boskos pellets, pelleted dried
sugar beet pulp, linseeds, oats, maize and various fruits and vegetables
were used by individual institutions. All giraffes, except for one, had
been housed separately overnight prior to culling. In an attempt to
estimate digesta retention post mortem, the animals had been offered
roughly 1 kg labelled silage and 0.5 kg labelled pellets at 18, 12 and
6 h before euthanasia. However, only 2 of the giraffes ate the labelled
feeds – the rest essentially fasted; therefore, no results on marker distri-
bution in the gastrointestinal tract were produced. Most giraffes were
fed their regular ration of concentrate pellets on the morning of
euthanasia. Therefore, two fasting times were estimated for each zoo
giraffe, one representing time since last access to roughage and one
representing time since last access to concentrate pellets. The average
durations for zoo giraffes were 14 ± 7 (range: 0.5–20 h) fasting time
for roughage and 2 ± 1 h for pellets, respectively.
2.1. Protocol for sampling and analyses
Dissections generally started within 30 min after euthanasia and the
sampling procedure in most cases took b1.5 h for wild-caught giraffes
and b2.5 h for captive giraffes. Giraffes were kept in an upright position
during transport to the dissection facility and during weighing. After
weighing, the animal was placed on its right side and the intestines
were ligated and removed. The forestomach was gently rolled over its
ventral edge onto the left side, as illustrated in Fig. 1. Great care
was taken to avoid any ‘kneading’ or pressure on any forestomach
region (in particular, the blindsacs and the reticulum), to minimize
mixing of rumen contents. The gastrointestinal tract sections were
separated as depicted in Fig. 2, after ligating each section. Before
collecting digesta samples, each section of the gastrointestinal tract,
i.e., the reticulorumen, omasum, abomasum, small intestine, cecum
and large intestine, was weighed full.
Digesta samples were collected for determining DM and fibre
concentrations in the dorsal rumen, ventral rumen, reticulum, omasum,
abomasum and rectum (Figs. 1 and 2), as described by Clauss et al.
(2009a). DM was determined by drying samples at 60 °C for 48 h until
constant weight; however, samples collected from 13 of the boma
giraffes were dried at 100 °C for 24 h. To quantify the effect of the drying
temperature on the results of the subsequent fibre analysis, samples
(n = 5) from one boma giraffe were manually divided into two
representative subsamples that were dried at 60 °C for 48 h or at
100 °C degrees for 24 h.
Prior to fibre analysis, the dried samples were ground to pass a 1 mm
screen (Retsch Ultra Centrifugal Mill ZM 200, Haan, Germany). The con-
tent of neutral detergent fibre (NDF) was determined using heat-stable
α-amylase and sodium sulphite in an ANKOM200 Fibre Analyzer
(ANKOM Technology, Macedon, NY, United States). After the NDF pro-
cedure, ANKOM sample bags containing the fibre content of each
 C. Sauer et al. / Comparative Biochemistry and Physiology, Part A 203 (2017) 69–76
Fig. 1. Schematic transects of a cranially viewed rumen changing position from upright to lying on its left side. The content depicted is highly stratified with a gas dome (white), a fibre mat
(straw-like), slurry beneath the mat (light grey) and a pool of dense particles at the bottom (dark grey). Dorsal (D) and ventral (V) samples of digesta were collected at the positions
indicated on the right most figure. This figure is for the illustration of the sampling method and is not representative for the rumen contents observed in giraffes.
sample were placed in a Dacron bag (38 μm pore size) and incubated in
rumen fistulated cows for 288 h to determine the fraction of indigestible
NDF (iNDF) as NDF residue after incubation and repeated NDF boiling
(Lund et al., 2007). Ash content in the iNDF fraction was determined
as the residue after dry-ashing the samples at 525 °C for 6 h, and both
NDF and iNDF concentrations were corrected for ash content. NDF and
iNDF results from samples dried at 100 °C were corrected using linear
regression equations derived from the identical subsamples dried at
60 °C or 100 °C, viz. NDF60 °C (in % of DM) = 0.61 × NDF100 °C (in % of
DM) + 9.17, R2 = 0.86 and iNDF60 °C (in % of DM) = 0.76 × iNDF100 °C
(in % of DM) − 1.84, R2 = 0.98.
At each sampling location (dorsal and ventral rumen, reticulum, oma-
sum, abomasum, rectum), a sample of approximately 30 g wet weight
was collected for determination of particle size distribution, and stored
frozen. These samples were analysed using a wet sieving technique
(Retsch AS Digit 200, Haan, Germany). Samples were soaked in 1 L of
water for a minimum of 12 h prior to analysis. Each sample was then
poured over a stack of 9 sieves with a linear reduction in mesh size
(16, 8, 4, 2, 1, 0.5, 0.25, 0.125 and 0.063 mm). The sample was mechani-
cally sieved for 10 min at a vibration amplitude of 2 mm and a water
throughput of 0.3 L/min. Particles not retained on the smallest screen
were not quantified. After sieving, the fraction of particles retained on
each sieve was carefully transferred to pre-weighed containers and
Fig. 2. Locations for sample collection in the stomachs. Locations for digesta sampling are
marked with a white square (□), while locations for mucosa sampling are marked with a
black circle (●). DR = dorsal rumen, AR = atrium ruminis, RE = reticulum, OM =
omasum, A = abomasum, VR = ventral rumen.
71
dried at 100 °C for 24 h. Mean particle size (in mm) of each sample was
calculated using the weighted average equation for the discrete mean
by Fritz et al. (2012), using an assumed maximum particle size of 32 mm.
Rumen fluid samples for VFA and NH3 analysis were collected at
each reticuloruminal location, i.e., dorsal rumen, ventral rumen and
reticulum. The fluid was filtered from the solid content of the sample
by the use of a filter bag with 0.5 mm pore size (Grade Blender Bags,
VWR, Denmark). Immediately after sampling, the filter bag was stored
at 5 °C for approximately 2 h before the fluid and the solid part were
fully separated. The concentration of NH3 in the rumen fluid was
analysed by making the sample alkaline with KOH, after which NH3
was determined by titration after distillation using the Kjeltec2400
system (Foss Analytical, Hillerød, Denmark). VFA concentrations were
determined by gas chromatography, as described by Jensen et al.
(1995) with some modifications (Canibe et al. 2007).
Samples for pH measurements were collected from the dorsal
rumen, ventral rumen, reticulum, omasum, abomasum, and rectum.
The pH value was measured using a portable pH meter (model IQ150,
IQ Scientific Instruments, Inc.). The time from euthanasia until measure-
ment of pH was noted for each sample. Reticulorumen and omasum pH
was usually measured within 20–25 min of each other. For the relatively
dry samples (omasum content and faeces), a small amount of
demineralized water was mixed into the sample to make it just
fluid enough for the pH meter to be able to measure pH.
After sampling, the content of each gastrointestinal section was
removed. The stomach sections were rinsed with water, squeezed and
allowed to drip-dry for at least 15 min before determining empty
weights, while the intestines were weighed without rinsing. The
wet weight of digesta in each gastrointestinal section was determined
by difference.
Mucosa samples from the dorsal rumen wall, the ventral rumen wall
and the wall of the Atrium ruminis were collected at the positions
marked with black circles on Fig. 2 and stored in 10% formalin until
dissection. A1.5 × 1.5 cm piece was cut from each mucosa sample
using a metal template. All papillae were subsequently cut from the
base of the square piece of mucosa and sorted according to size
(small, medium or large) based on visual judgement by the dissector
(CS). Number of papillae of each size was counted, and 10 papillae of
each size were randomly chosen and measured using a digital calliper.
For each papilla, length and width, determined at the midpoint
(i.e., at length/2), was measured. All measurements were recorded to
the nearest 0.01 mm. The surface enlargement factor (SEF) of the
rumen mucosa papillae was calculated as:
ðno:of papillae  MPSAÞ þ base surface
SEF ¼ ;
base surface
where MPSA = mean papillae surface area measured in cm2
(i.e., 2 × mean papillae height (cm) × mean papillae width (cm)), and
base surface = area of the piece of mucosa dissected (i.e., 2.25 cm2).
72 C. Sauer et al. / Comparative Biochemistry and Physiology, Part A 203 (2017) 69–76
2.2. Statistical analyses
For each measurement, repeated measures ANOVA were used to
determine differences between boma and zoo giraffes by models that
included fasting time or log-transformed body mass as a covariate
and origin (boma/zoo) as a between-subjects factor (Table S1; with
stepwise exclusion of non-significant factors), with Sidak post hoc
tests to adjust for multiple comparisons. Similarly, simple repeated
measures ANOVA with Sidak post hoc tests were used to determine
differences between forestomach locations. Linear regressions of log-
transformed data served to identify scaling relationships with body
mass. Additional pair-wise comparisons were made with t-tests. All
statistical analyses were performed using the statistical software SPSS
22.0 (IBM, Armonk, NY). The significance level α was set to 0.05.
3. Results and discussion
Measurements not directly related to rumen contents stratification
(such as contents weights or measures in the abomasum and lower
digestive tract) are given in Appendix A (incl. Figs. S1–S7).
3.1. Digesta dry matter content
The difference in DM concentration between the dorsal and ventral
rumen regions was not significant (boma: P = 0.290, zoo: P = 0.637),
while the reticular DM concentration was lower than in the rumen in
boma (P ≤ 0.020) but not zoo giraffes (P ≥ 0.063). DM concentration in
the omasum was higher than in the RR (boma: P b 0.001, zoo: P =
0.001, Fig. 3A). The absence of a difference in DM concentration
between the dorsal and ventral part of the rumen indicates
homogenous rumen content in the giraffe, as expected.
When compared to other ruminants, i.e., addax (Addax
nasomaculatus), bison (Bison bison), and moose (Alces alces) (Clauss
et al., 2009a), the lack of difference in DM concentration between the
dorsal and ventral rumen in giraffes is similar to findings in moose, an-
other browsing ruminant (Fig. 3B), whereas the grazing, ‘cattle-type’
ruminants showed a clear difference in dorsal and ventral DM concen-
tration, as evidence of a stratified rumen content. Overall, DM concen-
trations were lower in this study than previously reported for giraffes
(reticulorumen: 13.5–13.8%, n = 9) (Maloiy et al., 1982; Clemens
and Maloiy, 1983), which is probably due to the long fasting times in
this study and to some extent to dietary differences between studies,
both of which have been documented to affect digesta DM (Hummel
et al., 2009).
Fig. 3. Dry matter concentration of digesta samples collected along the digestive tract of the giraffe and other ruminant species. DR = dorsal rumen, VR = ventral rumen, RE = reticulum,
OM = omasum. Bars represent means and standard deviations. A) Dry matter concentration in giraffe digesta. Samples from 16 boma and 9 zoo giraffes. No common superscripts denote
significant differences between sampling locations (small letters for boma, capital letters for zoo). B) Dry matter concentration in forestomach digesta of boma giraffes (this study),
compared to addax (Addax nasomaculatus), bison (Bison bison), and moose (Alces alces) (from Clauss et al., 2009a). No common superscripts denote significant differences between
sampling locations within a species.
3.2. Digesta particle size
There was no difference in mean particle size between boma
and zoo giraffes at the reticuloruminal sampling locations (Fig. 4A,
P = 0.391–0.993), whereas zoo giraffes had larger particles than
boma giraffes in the omasum (1.28 ± 0.35 mm vs. 0.81 ± 0.30 mm,
P = 0.016) and in the faeces (0.85 ± 0.12 mm vs. 0.57 ± 0.10 mm,
P = 0.001). Larger faecal particles in captive versus free-ranging
giraffes have previously been reported (0.75 versus 0.35 mm, respec-
tively) (Hummel et al., 2008). This difference between captive and
free-ranging animals have also been found in another browser, the
moose, whereas no similarly extreme difference is evident in grazing
ruminants (Hummel et al., 2008; Lechner et al., 2010). This has been
linked to the fact that diets fed to captive browsers likely differ more
from natural diets than diets fed to captive grazers, and to reports of
excessive tooth wear in captive giraffes (Clauss et al., 2007). Worn
down teeth reducing chewing efficiency in captive individuals
should result in larger particles in the reticulorumen as well.
Although boma giraffes generally had longer fasting times, many of
them had been under anaesthesia and thus unable to ruminate for up
to 8 h prior to euthanasia, whereas the zoo giraffes were able to
ruminate up until the time of euthanasia. The extended rumination in
zoo giraffes may have counterbalanced any original differences in
reticulorumen particle size when compared to boma giraffes. There
was no effect of fasting time on mean particle size (P = 0.655). This
finding may be a result of the combination of shorter rumination time
and smaller initial particle size in boma individuals versus longer
rumination time and larger initial particle size in captive individuals,
cancelling the effect of fasting time.
Within individual giraffes, mean particle size did not differ between
dorsal and ventral rumen (boma: P = 1.000, zoo: P = 0.724), again
supporting the hypothesis of unstratified rumen content in giraffe.
In boma giraffe, there were no differences between rumen and
reticulum mean particle size (dorsal rumen-reticulum: P = 0.082,
ventral rumen-reticulum: P = 0.164), and the omasum contained
smaller particles than any preceding location (P ≤ 0.002). In zoo giraffes,
although a numerical reduction of mean particle size was observed from
the reticulorumen to the omasum (Fig. 4A), it was not significant
(P ≥ 0.132).
When compared to other species of ruminants (Clauss et al., 2009a),
giraffes had remarkably smaller particles in all reticulorumen locations
(Fig. 4B). In addition, the magnitude of particle size reduction from
the reticulorumen to the omasum, indicating selective retention of
large particles in the reticulorumen, was much lower than in the other
three species, yet the particle size in the omasum was similarly small.
This is likely due to two reasons. Firstly, the animals had an unknown
 C. Sauer et al. / Comparative Biochemistry and Physiology, Part A 203 (2017) 69–76
Fig. 4. Mean particle size of digesta samples collected along the digestive tract of the giraffe and other ruminant species. DR = dorsal rumen, VR = ventral rumen, RE = reticulum, OM =
omasum. Bars represent means and standard deviations. a) Mean particle size of giraffe digesta in samples from 10 boma and 5 zoo giraffes. No common superscripts denote significant
differences between sampling locations (small letters: boma, capital letters: zoo); *denotes difference between boma and zoo. b) Mean particle size in forestomach digesta of giraffes (this
study), compared to addax (Addax nasomaculatus), bison (Bison bison), and moose (Alces alces) (from Clauss et al., 2009a). No common superscripts denote significant differences between
sampling locations within a species.
proportion of pelleted (and hence, ground) feed in their diet. Secondly,
the majority of both the boma and zoo giraffes in this study had been
fasted for a substantial period of time, while continuing to ruminate
and thereby reducing particle size (McLeod and Minson, 1988).
Nevertheless, differences in particle size between dorsal and ventral
rumen content have been documented in cattle for up to 24 h after
feeding (Hummel et al., 2009) and in goats 12 h after feeding (Clauss
et al., 2016), suggesting that the absence of particle size stratification
in the giraffes of the present study may indeed indicate an absence of
reticulorumen contents stratification. Although 75% of particle size re-
duction has been attributed to masticatory activity (initial chewing
and rumination; McLeod and Minson, 1988), microbial fermentation
of the feed particles will reduce the size as well (McLeod and Minson,
1988; Krämer et al., 2013). Thus, the reduction in particle size from
the omasum to the faeces observed in this study (Fig. S3) indicates
considerable microbial activity in the hindgut of the giraffe; however,
the extent of this fermentation remains to be investigated.
3.3. Digesta fibre content
The average iNDF:NDF ratio in the whole digestive tract was signifi-
cantly higher in boma than in zoo giraffes (0.65 ± 0.08 vs. 0.46 ± 0.12,
P ≤ 0.008 in all location-wise comparisons, Fig. S4). However, origin was
not significant in the overall model, but there was a significant effect of
fasting time (P = 0.048). Fasting time differed significantly in this com-
parison (887 ± 632 min in boma vs. 114 ± 74 min in zoo, P b 0.001),
and was positively related to the iNDF:NDF ratio, indicating that more
time for digestion led, expectedly, to a higher proportion of indigestible
material. The higher iNDF:NDF ratio observed in boma giraffes is thus a
consequence of their longer fasting time. Another contributing factor
could have been differences in diets, as the boma diet was likely less
digestible compared to the zoo diet, i.e., the boma diet probably had a
higher iNDF concentration to start with. Within individual giraffes,
there was no difference in iNDF:NDF ratio between the forestomach
locations (boma: P = 0.069–1.000; zoo: P = 0.968–1.000). The lack of
difference between the dorsal and ventral rumen indicates homogenous
rumen content.
3.4. Digesta concentrations of VFA and NH3
Zoo giraffes generally had numerically higher concentrations of total
VFAs (P = 0.064–0.090), of acetic acid (P = 0.064–0.097), propionic
acid (P = 0.062–0.088), butyric acid (P = 0.026–0.045) and NH3
(P = 0.025–0.039) in the rumen and reticulum compared to boma
giraffes (Fig. S5), again indicating more digestible diets and shorter
fasting times. There was no difference between sampling locations for
73
concentrations of total VFA, any individual VFAs or NH3 (P ≥ 0.125,
Fig. S5). Boma giraffes had a higher mean proportion of acetic
acid (0.73 ± 0.02 vs. 0.69 ± 0.05, P = 0.023) and lower proportion of
butyric acid (0.10 ± 0.01 vs. 0.14 ± 0.02, P = 0.001) than zoo giraffes,
while the proportion of propionic acid was the same (0.17 ± 0.02 vs.
0.18 ± 0.03, P = 0.487). To summarise, the molar proportions of
acetic:propionic:butyric acid found is this study were 73:17:10 for
boma giraffes and 69:18:14 for zoo giraffes. Ratios of 76:14:9 and
73:14:13 have previously been documented in wild giraffes (Maloiy
et al., 1982, n = 6; Clemens and Maloiy, 1983, n = 3). The difference
in the proportion of propionic acid could be due to the inclusion of
pelleted feeds in the present study. Commercial dairy cows have an
average ratio of 63:22:11 (101 different diets, Morvay et al., 2011).
The average reticuloruminal concentration of NH3 in the present
study (boma giraffes: 18.5 ± 6.9 mg/100 mL and zoo giraffes: 33.7 ±
14.3 mg/100 mL) was in the range of values previously reported for 6
free-ranging giraffes of 24.6 ± 2.1 mg/100 mL (Maloiy et al., 1982),
while the total VFA concentration (boma giraffes: 47.8 ± 18.8 mmol/L
and zoo giraffes: 82.7 ± 42.7 mmol/L) in this study was lower than pre-
viously reported for giraffes (Maloiy et al., 1982, 158.3 ± 3.5 mmol/L;
Clemens and Maloiy, 1983, 106.4 ± 10.6 mmol/L). Factors like differ-
ences in diet composition (Hummel et al., 2006), amounts fed and
time passed since last feeding (Brask et al., 2015), greatly influence
the concentration of fermentation products in the rumen of domestic
ruminants, and are likely the cause of the large variation observed
among giraffes in this study and also when comparing between studies.
Differences in sampling protocols presumably contribute as well, with
variation in lag time from death to sampling, the presence/absence of
a fixation agent (e.g., H2SO4 used by Maloiy et al., 1982; Clemens and
Maloiy, 1983) and potential differences in analytical procedures
between studies.
3.5. Digesta pH
The pH in the dorsal rumen, ventral rumen and omasum decreased
(P = 0.022, 0.030 and 0.033, respectively) with time since euthanasia,
while abomasum pH increased (P b 0.001, Fig. S6A). Fasting time had
a (positive) effect only on the pH of the dorsal and ventral rumen
(P = 0.022 and 0.049, respectively). When accounting for the longer
time since euthanasia in zoo giraffes, there was no difference between
boma and zoo giraffes in pH of the ventral rumen (P = 0.062) and
abomasum (P = 0.277), but a difference was found in the dorsal
rumen (P = 0.036) and the omasum (P = 0.021). The pH of the
reticulum was unaffected by both time since euthanasia (P = 0.248)
and giraffe origin (P = 0.170). Similar to this study, Ritz et al. (2013)
observed a decrease of rumen pH with longer time since euthanasia in
74 C. Sauer et al. / Comparative Biochemistry and Physiology, Part A 203 (2017) 69–76
roe deer (Capreolus capreolus), suggesting a continuation of microbial
activity after death while absorption of VFA terminates at the time of
euthanasia.
Within individual giraffes, there was no difference in pH between
any forestomach locations (Fig. 5, boma: P = 0.084–0.998, zoo: P =
0.628–1.000), and thus no indication of a higher fermentation activity
at either site. The pH measured in this study was comparable to values
previously reported for reticuloruminal and abomasal content (pH of
6.5 and 3.6, respectively) of six free-ranging giraffes (Maloiy et al.,
1982). The numerically lower pH in the omasum as compared to the
rumen (Fig. 5) resembles findings reported in both red deer (Cervus
elaphus) and fallow deer (Dama dama), with rumen – omasum
differences of 0.7 and 0.5, respectively (Prins et al., 1972). The pH in
the reticulum was unaffected by time since euthanasia in this study –
an observation made in roe deer as well (Ritz et al., 2013). Perhaps
the greater amount of buffering saliva in the reticulum makes this
location less sensitive to time effects compared to other forestomach
locations.
3.6. Rumen papillation
There was a negative relationship between body mass and papillae
density (number of papillae per cm2) (P = 0.003), with a scaling
exponent that included − 1.0 in the 95% confidence interval for the
dorsal and the ventral rumen, and came close to that value (− 0.95)
for the Atrium ruminis (Fig. S7). Accounting for body mass, papillae den-
sity was lower in zoo giraffes than in boma giraffes (P = 0.022, Table 1).
In boma giraffes, papillae density in the ventral rumen was lower
than both in the dorsal rumen (P b 0.001) and the Atrium ruminis
(P = 0.040; Table 1). In zoo giraffes, papillae density in the Atrium
ruminis was higher than in the ventral rumen (P = 0.005) but not in
the dorsal rumen (P = 0.081; Table 1).
The average area of individual papilla of each location was positively
correlated to body mass (all P b 0.001), with a scaling exponent that al-
ways included 1.0 in the 95% confidence interval (Fig. S7). Accounting
for body mass, papilla area was much greater in zoo than in boma
giraffes (P b 0.001, Table 1). In zoo giraffes, there was no difference in
average papilla area between sampling locations (P = 0.688–0.850),
whereas papillae in the ventral rumen were larger on average than
papillae in the dorsal rumen in boma giraffes (P = 0.007, Table 1).
There was no significant relationship between body mass and SEF
(P = 0.261), with a scaling exponent that always included 0 in the
95% confidence interval (Fig. S7). SEF was greater in zoo than in boma
giraffes (P = 0.014, Table 1). In spite of numerical differences
(Table 1), there were no significant differences between forestomach
Fig. 5. pH in digesta samples collected in the forestomach of 6 boma and 7 zoo giraffes.
DR = dorsal rumen, VR = ventral rumen, RE = reticulum, OM = omasum. Bars
represent mean values with standard deviation; *denotes differences between boma
and zoo (accounting for fasting time and time between euthanasia and measurement).
There were no significant differences between forestomach locations in neither boma
nor zoo giraffes.
Fig. 6. Relation between dorsal to ventral difference in rumen digesta dry matter
concentration and relative SEF Dorsal rumen (in % SEF Atrium ) in giraffes (this study)
and other ruminants (from Codron and Clauss, 2010). Each point represents the
mean of a species.
locations for the SEF within boma (P = 0.085–0.996) or zoo giraffes
(P = 0.147–0.992).
Characteristics of the rumen papillae are typically influenced by both
body size and diet (e.g. Josefsen et al., 1996; Wang et al., 2009). In this
study, both effects were apparent. To our knowledge, the relationship
between the intraruminal papillation pattern and body mass has hardly
been investigated quantitatively, and the fact that the scaling exponents
of papillae density and papilla area cancel out each other, resulting in no
scaling of SEF, is a very interesting side finding. The observation that
older, and hence larger animals have lower papillae densities and larger
papillae was reported repeatedly (Berg et al., 1986; Josefsen et al., 1996;
Mathiesen et al., 2000). A parsimonious explanation could be that
rumen papillae are already developed in the embryo (e.g. Franco et al.,
2011), and that their number is comparatively fixed. Notably, the
rumen of neonate ruminants is homogenously covered with papillae,
even at locations that have no papillae in the adult forms of some
species such as the dorsal rumen or the ruminal pillars (Hofmann,
1973). Growth of both the rumen wall and the papillae necessarily
leads to a reduction of the number of papillae per defined unit of area
(such as cm2), while their total number may stay comparatively stable
(Berg et al., 1986). Variation in papillae density between animals of sim-
ilar age or size might then mainly reflect conditions that prevent devel-
opment of all preformed papillae. This explanation matches
observations that dietary manipulations that increase papilla size or pa-
pillae surface often do not increase papillae density (Shen et al., 2004;
Wang et al., 2009; Xu et al., 2009). In our study, zoo giraffes were larger
than boma giraffes (mean body mass of 687 ± 318 vs. 475 ± 88 kg) and
had a lower density of papillae, but with each individual papilla having a
larger surface area, which resulted in overall higher SEF. The fact that
the difference between zoo and boma giraffes mostly persisted even
when correcting for body mass indicates an additional dietary effect,
with a more digestible diet in zoo giraffes as mentioned repeatedly
above.
When Clauss et al. (2009b) evaluated SEF data from 59 ruminant
species, it was noted that the dorsal rumen-SEF value of 24 found in
wild giraffes by Hofmann (1973) was unusually high (Table 1; possibly
because the smaller papillae were not included in the measurement), as
11.0 was the second highest dorsal rumen-SEF found in the entire
dataset (for grey duiker, Sylvicapra grimmia). The dorsal rumen-SEF
found in this study (average of 13.5) appear more moderate; though
still the highest of all ruminant species.
When dorsal rumen-SEF and ventral rumen-SEF were expressed in %
of the SEF of the Atrium ruminis (termed relative SEF), both were numer-
ically higher in boma than zoo giraffes (Table 1), though not statistically
different (P = 0.108 and P = 0.438, respectively). The relative SEF
differed between the dorsal and the ventral rumen in boma giraffes
 C. Sauer et al. / Comparative Biochemistry and Physiology, Part A 203 (2017) 69–76 75

Table 1
Characteristics of rumen mucosa papillae in boma and zoo giraffes.

Boma giraffes Zoo giraffes

This study1 H. (1973)2 This study1 H.M. (1988)1

Measure Location
n = 12 n=6 n=7 n=2

Papillae density DR 80.0 ± 21.6a 24 40.7 ± 26.9ab 17/15

(papillae/cm2) AR 74.9 ± 17.7a 26 54.9 ± 24.0a 9/15
VR 52.8 ± 20.2b 19 34.9 ± 16.1b 14/12

Papillae area DR 16.1 ± 4.1b – 41.3 ± 14.9 –

(mm2) AR 18.1 ± 5.9ab – 44.6 ± 16.8 –
VR 19.5 ± 3.8a – 51.1 ± 32.2 –

SEF DR 13.5 ± 3.7 24 15.1 ± 4.1 2.7/1.9

AR 14.1 ± 3.1 30 22.5 ± 6.3 4.95/4.59
VR 11.4 ± 5.1 18 15.8 ± 5.5 2.47/2.74

Relative SEF DR 100.1 ± 33.2 80.0 73.5 ± 32.5 53.9/41.4

(% of SEFAR) VR 87.2 ± 44.0 60.0 72.7 ± 24.2 49.9/59.7

H. (1973) = Hofmann (1973), H.M. (1988) = Hofmann and Matern (1988), DR = dorsal rumen, AR = atrium ruminis, VR = ventral rumen, SEF = surface enlargement factor.
No common superscripts denote significant differences between sampling locations within a group.
1
Three sizes of papillae were recognized and all were included in the papillae count.
2
Three sizes of papillae were recognized, but only the two largest were included in the papillae count.

(P = 0.033), but not in zoo giraffes (P = 0.957). No effect of body mass
could be demonstrated for the relative SEF (P = 0.603). The relative SEF
was negatively related to the difference in dry matter concentration be-
tween the contents of the dorsal and the ventral rumen (Fig. 6), suggest-
ing that these two measures of rumen contents stratification are linked
as expected. The relative SEF of boma giraffes were numerically much
closer to 100% than in zoo giraffes, indicating slightly more stratified
rumen content in the zoo giraffes. This difference matches previous ob-
servations in wild and captive giraffes (Table 1), and supports the as-
sumption that feeding regimes in captivity might lead to more
stratified rumen content in giraffe (Hummel and Clauss, 2006). This
finding might be related to the difficulty of providing captive giraffe
with adequate amounts and ‘quality’ of roughage. Lucerne hay is consid-
ered an indispensable diet component for captive giraffes, due to the fact
that these animals require a source of structurally effective fibre, yet
usually do not ingest grass hay in relevant amounts but typically accept
lucerne hay more readily. However, maintaining giraffe on a diet of lu-
cerne hay alone may still be problematic due to low ad libitum intakes
(Hatt et al., 2005). As a dicot, lucerne shares some characteristics with
browse, such as a higher degree of lignification than grass, as well as
similar fermentation and physical fractionation patterns (Troelsen and
Campbell, 1968; Gussek et al., 2016). Nevertheless, cattle fed only lu-
cerne hay still display a pronounced rumen content stratification
(Hummel et al., 2009). Possibly, only a dietary regime consisting pre-
dominantly of browse, which is very difficult to achieve in captivity,
would result in the particularly homogenous papillation pattern ob-
served in free-ranging animals.
4. Conclusion
The physical characteristics of the rumen content of boma and zoo
giraffes were investigated in this study. No difference between the
dorsal and ventral rumen content was evident in any measurement
(DM, particle size, iNDF:NDF, VFA, NH3 or pH), thus indicating an un-
stratified rumen content in the giraffe. This observation was further
supported by the finding of an even rumen papillation pattern, as evi-
denced by high relative SEF of both the dorsal and ventral rumen, in
boma giraffes. A numerical difference in relative SEF indicated a slightly
less homogenous rumen content in zoo giraffes, but this finding could
not be substantiated by physical characteristics of the rumen content.
Nevertheless, it indicates that the diet these captive animals received
in the last weeks before sampling, and hence most likely during
their whole life in captivity, induces some degree of rumen content
stratification not observed in animals from the wild.
Acknowledgements
The authors would like to thank the Danish Cardiovascular Giraffe
Research Program for the opportunity to collect data from a large num-
ber of wild-caught giraffes. Further, the authors greatly appreciate the
donations of captive giraffes from a number of Danish zoos to the
study. The assistance of Stine Kugle, Helle Hydeskov, Tanja N. Gade
and Mari-Ann Da Silva during the sampling collections was invaluable.
Morten Poulsen from AU Foulum kindly contributed his expertise on
running VFA analyses on the samples. Peder Nørgaard at Copenhagen
University is gratefully recognized for allowing the use of his wet
sieving equipment for the particle size analyses. The help of the lab tech-
nicians at AU Foulum, in particular Inger Østergaard and Hanne V. Berg,
is much appreciated. We thank three anonymous reviewers for
their comments.
Appendix A. Supplementary material (incl. Figs. S1–S7)
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.cbpa.2016.08.033.
References
Berg, R., Rieger, D.C., Enany, E.I., 1986. Anatomical studies into the ruminal mucosa of
pygmy goats in relation to their age. Arch. Anim. Nutr. 36, 611–616.
Brask, M., Weisbjerg, M.R., Hellwing, A.L.F., Bannink, A., Lund, P., 2015. Methane
production and diurnal variation measured in dairy cows and predicted from
fermentation pattern and nutrient or carbon flow. Animal 9, 1795–1806.
Canibe, N., Højberg, O., Badsberg, J.H., Jensen, B.B., 2007. Effect of feeding fermented liquid
feed and fermented grain on gastrointestinal ecology and growth performance in pig-
lets. J. Anim. Sci. 85, 2959–2971.
Clauss, M., Lechner-Doll, M., Streich, W.J., 2002. Faecal particle size distribution in captive
wild ruminants: an approach to the browser/grazer dichotomy from the other end.
Oecologia 131, 343–349.
Clauss, M., Lechner-Doll, M., Streich, W.J., 2003. Ruminant diversification as an adaptation
to the physicomechanical characteristics of forage. A reevaluation of an old debate
and a new hypothesis. Oikos 102, 253–262.
Clauss, M., Franz-Odendaal, T.A., Brasch, J., Castell, J.C., Kaiser, T.M., 2007. Tooth wear in
captive giraffes (Giraffa camelopardalis): mesowear analysis classifies free-ranging
specimens as browsers but captive ones as grazers. J. Zoo Wildl. Med. 38, 433–445.
Clauss, M., Kaiser, T., Hummel, J., 2008. The Morphophysiological Adaptations of Browsing
and Grazing Mammals. In: Gordon, I.J., Prins, H.H.T. (Eds.), The Ecology of Browsing
and Grazing. Springer, Heidelberg, pp. 47–88.
Clauss, M., Fritz, J., Bayer, D., Nygren, K., Hammer, S., Hatt, J.-M., Südekum, K.-H., Hummel,
J., 2009a. Physical characteristics of rumen contents in four large ruminants of differ-
ent feeding type, the addax (Addax nasomaculatus), bison (Bison bison), red deer
(Cervus elaphus) and moose (Alces alces). Comp. Biochem. Physiol. A 152, 398–406.
Clauss, M., Hofmann, R.R., Fickel, J., Streich, W.J., Hummel, J., 2009b. The intraruminal
papillation gradient in wild ruminants of different feeding types: implications for
rumen physiology. J. Morphol. 270, 929–942.
Clauss, M., Hume, I.D., Hummel, J., 2010. Evolutionary adaptations of ruminants and their
potential relevance for modern production systems. Animal 4, 979–992.
76 C. Sauer et al. / Comparative Biochemistry and Physiology, Part A 203 (2017) 69–76
Clauss, M., Fritz, J., Tschuor, A., Braun, U., Hummel, J., Codron, D., 2016. Dry matter and
digesta particle size gradients along the goat digestive tract on grass and browse
diets. J. Anim. Physiol. Anim. Nutr. http://dx.doi.org/10.1111/jpn.12505 (online).
Clemens, E.T., Maloiy, G.M.O., 1983. Digestive physiology of East African wild ruminants.
Comp. Biochem. Physiol. A 76, 319–333.
Codron, D., Clauss, M., 2010. Rumen physiology constrains diet niche: linking digestive
physiology and food selection across wild ruminant species. Can. J. Zool. 88,
1129–1138.
Dittmann, M.T., Hummel, J., Hammer, S., Arif, A., Hebel, C., Müller, D.H.W., Fritz, J., Steuer,
P., Schwarm, A., Kreuzer, M., Clauss, M., 2015. Digesta retention in gazelles in
comparison to other ruminants: evidence for taxon-specific rumen fluid throughput
to adjust digesta washing to the natural diet. Comp. Biochem. Physiol. A 185, 58–68.
Franco, A., Masot, J., Redondo, E., 2011. Ontogenesis of the rumen: a comparative analysis
of the Merino sheep and Iberian red deer. Anim. Sci. J. 82, 107–116.
Fritz, J., Streich, W.J., Schwarm, A., Clauss, M., 2012. Condensing results of wet sieving
analyses into a single data: a comparison of methods for particle size description.
J. Anim. Physiol. Anim. Nutr. 96, 783–797.
Gussek, I., Große-Brinkhaus, C., Hummel, J., Südekum, K.-H., 2016. Chemical composition
and fermentation characteristics of feedstuffs for giraffes (Giraffa camelopardalis) in
German zoos. J. Anim. Feed Sci. 25, 134–144.
Hatt, J.-M., Schaub, D., Wanner, M., Wettstein, H.R., Flach, E.J., Tack, C., Hässig, M.,
Ortmann, S., Hummel, J., Clauss, M., 2005. Energy and fibre intake in a group of
captive giraffe (Giraffa camelopardalis) offered increasing amounts of browse.
J. Veterinary Med. Ser. A 52, 485–490.
Heywood, J.J.N., 2010. Functional anatomy of bovid upper molar occlusal surfaces with
respect to diet. J. Zool. 281, 1–11.
Hofmann, R.R., 1973. The Ruminant Stomach. East African Literature Bureau, Nairobi.
Hofmann, R.R., 1989. Evolutionary steps of ecophysiological adaptation and diversification
of ruminants: a comparative view of their digestive system. Oecologia 78, 443–457.
Hofmann, R.R., Matern, B., 1988. Changes in gastrointestinal morphology related to
nutrition in giraffes (Giraffa camelopardalis): a comparison of wild and zoo
specimens. Int. Zoo Yb. 27, 168–176.
Hofmann, R.R., Streich, W.J., Fickel, J., Hummel, J., Clauss, M., 2008. Convergent evolution
in feeding types: salivary gland mass differences in wild ruminant species. J. Morphol.
269, 240–257.
Hummel, J., Clauss, M., 2006. Feeding, EAZA Husbandry and Management Guidelines for
Giraffa camelopardalis. Burger's Zoo, Arnhem, pp. 29–61.
Hummel, J., Südekum, K.-H., Streich, W.J., Clauss, M., 2006. Forage fermentation patterns and
their implications for herbivore ingesta retention times. Funct. Ecol. 20, 989–1002.
Hummel, J., Fritz, J., Kienzle, E., Medici, E.P., Lang, S., Zimmermann, W., Streich, W.J.,
Clauss, M., 2008. Differences in fecal particle size between free-ranging and captive
individuals of two browser species. Zoo Biol. 27, 70–77.
Hummel, J., Südekum, K.-H., Bayer, D., Ortmann, S., Hatt, J.-M., Streich, W.J., Clauss, M.,
2009. Physical characteristics of reticuloruminal contents of cattle in relation to
forage type and time after feeding. J. Anim. Physiol. Anim. Nutr. 93, 209–220.
Hummel, J., Hammer, S., Hammer, C., Ruf, J., Lechenne, M., Clauss, M., 2015. Solute and
particle retention in a small grazing antelope, the blackbuck (Antilope cervicapra).
Comp. Biochem. Physiol. A 182, 22–26.
Jensen, M.T., Cox, R.P., Jensen, B.B., 1995. Microbial production of skatole in the hind gut of
pigs given different diets and its relation to skatole deposition in backfat. Anim. Sci.
61, 293–304.
Josefsen, T.D., Aagnes, T.H., Mathiesen, S.D., 1996. Influence of diet on the morphology of
the ruminal papillae in reindeer calves (Rangifer tarandus). Rangifer 16, 119–128.
Kaiser, T.M., Fickel, J., Streich, W.J., Hummel, J., Clauss, M., 2010. Enamel ridge alignment in
upper molars of ruminants in relation to their natural diet. J. Zool. 281, 12–25.
Krämer, M., Nørgaard, P., Lund, P., Weisbjerg, M.R., 2013. Particle size alterations
of feedstuffs during in situ neutral detergent fiber incubation. J. Dairy Sci. 96,
4601–4614.
Lechner, I., Barboza, P., Collins, W., Fritz, J., Günther, D., Hattendorf, B., Hummel, J.,
Südekum, K.-H., Clauss, M., 2010. Differential passage of fluids and different-
sized particles in fistulated oxen (Bos primigenius F. taurus), muskoxen (Ovibos
moschatus), reindeer (Rangifer tarandus) and moose (Alces alces): rumen particle
size discrimination is independent from contents stratification. Comp. Biochem.
Physiol. A 155, 211–222.
Leuthold, B.M., Leuthold, W., 1972. Food habits of giraffe in Tsavo National Park, Kenya. E.
Afr. Wildl. J. 10, 129–142.
Lund, P., Weisbjerg, M.R., Hvelplund, T., 2007. Digestible NDF is selectively retained in
the rumen of dairy cows compared to indigestible NDF. Anim. Feed Sci. Technol. 134, 1–17.
Maloiy, G.M.O., Clemens, C.T., Kamau, J.M.Z., 1982. Aspects of digestion and in vitro rumen
fermentation rate in six species of East African wild ruminants. J. Zool. (Lond.) 197,
345–353.
Mathiesen, S.D., Haga, Ø.E., Kaino, T., Tyler, N.J.C., 2000. Diet composition, rumen
papillation and maintenance of carcass mass in female Norwegian reindeer
(Rangifer tarandus tarandus) in winter. J. Zool. 251, 129–138.
McLeod, M.N., Minson, D.J., 1988. Large particle breakdown by cattle eating ryegrass and
alfalfa. J. Anim. Sci. 66, 992–999.
Morvay, Y., Bannink, A., France, J., Kebreab, E., Dijkstra, J., 2011. Evaluation of models to
predict the stoichiometry of volatile fatty acid profiles in rumen fluid of lactating
Holstein cows. J. Dairy Sci. 94, 3063–3080.
Müller, D.W.H., Bingaman Lackey, L., Streich, W.J., Fickel, J., Hatt, J.-M., Clauss, M., 2011.
Mating system, feeding type and ex-situ conservation effort determine life
expectancy in captive ruminants. Proc. R. Soc. B 278, 2076–2080.
Pellew, R.A., 1984. The feeding ecology of a selective browser, the giraffe (Giraffa
camelopardalis tippelskirchi). J. Zool. 202, 57–81.
Prins, R.A., Hungate, R.E., Prast, E.R., 1972. Function of the omasum in several ruminant
species. Comp. Biochem. Physiol. A 43, 155–163.
Renecker, L.A., Hudson, R.J., 1990. Digestive kinetics of moose, wapiti and cattle. Anim.
Prod. 50, 51–61.
Ritz, J., Hofer, K., Hofer, E., Hackländer, K., Immekus, D., Codron, D., Clauss, M., 2013.
Forestomach pH in hunted roe deer (Capreolus capreolus) in relation to forestomach
region, time of measurement, supplemental feeding, and a comparison among wild
ruminant species. Eur. J. Wildl. Res. 59, 505–517.
Sauer, C., Bertelsen, M.F., Lund, P., Weisbjerg, M.R., Clauss, M., 2016. Quantitative
macroscopic anatomy of the giraffe (Giraffa camelopardalis) digestive tract. Anat.
Histol. Embryol. http://dx.doi.org/10.1111/ahe.12201 (online).
Seeber, P.A., Ndlovu, H.T., Duncan, P., Ganswindt, A., 2012. Grazing behaviour of the giraffe
in Hwange National Park, Zimbabwe. Afr. J. Ecol. 50, 247–250.
Shen, Z., Seyfert, H.M., Löhrke, B., Schneider, F., Zitnan, R., Chudy, A., Kuhla, S., Hammon,
H.M., Blum, J.W., Martens, H., Hagemeister, H., Voigt, J., 2004. An energy-rich diet
causes rumen papillae proliferation associated with more IGF type 1 receptors and
increased plasma IGF-1 concentrations in young goats. J. Nutr. 134, 11–17.
Smerup, M., Damkjær, M., Brøndum, E., Baandrup, U.T., Kristiansen, S.B., Nygaard, H.,
Funder, J., Aalkjær, C., Sauer, C., Buchanan, R., Bertelsen, M.F., Østergaard, K.,
Grøndahl, C., Candy, G., Hasenkam, J.M., Secher, N.H., Bie, P., Wang, T., 2016. The
thick left ventricular wall of the giraffe heart normalises wall tension, but limits
stroke volume and cardiac output. J. Exp. Biol. 219, 457–463.
Troelsen, J.E., Campbell, J.B., 1968. Voluntary consumption of forage by sheep and its relation
to the size and shape of particles in the digestive tract. Anim. Prod. 10, 289–296.
Tschuor, A., Clauss, M., 2008. Investigations on the stratification of forestomach contents
in ruminants: an ultrasonographic approach. Eur. J. Wildl. Res. 54, 627–633.
Wang, Y.H., Xu, M., Wang, F.N., Yu, Z.P., Yao, J.H., Zan, L.S., Yang, F.X., 2009. Effect of dietary
starch on rumen and small intestine morphology and digesta pH in goats. Livest. Sci.
122, 48–52.
Welch, J.G., 1982. Rumination, particle size and passage from the rumen. J. Anim. Sci. 54,
885–894.
Xu, M., Dong, Y., Du, S., Hao, Y.S., Wang, Y.H., Wang, F.N., Yao, J.H., 2009. Effect of corn
particle size on mucosal morphology and digesta pH of the gastrointestinal tract in
growing goats. Livest. Sci. 123, 34–37.